Daily Archives: February 5, 2009

Managing The Purchasing Factory


Today’s guest post is from Pierre Mitchell, Director, Procurement Research and Advisory for The Hackett Group.

Dave Nelson, the CPO from John Deere, co-authored a book titled “The Purchasing Machine“. The book was good, but never explained the meaning of the title. It did however get us thinking about the analogy of a factory to a Procurement function, and how Procurement can apply Lean Manufacturing principles to its operations.

Many companies are currently implementing Six Sigma methodologies, and both Lean and Six Sigma emphasize a focus on the customer and the elimination of waste. Six Sigma’s “DMAIC” methodology can very easily be applied (and is being applied at some progressive organizations):

  • Defining the needs of procurement’s internal customers,
  • Measuring the criteria of success (e.g., supply assurance, savings, supplier innovation, etc.),
  • Analyzing the current situation (e.g., too many suppliers, too many ways to buy, etc.),
  • Improving the processes (i.e., the “opportunity identification” step in a sourcing methodology), and
  • Controlling processes to “hold the gains” (e.g., contract compliance) through fail-safe processes.

This is foundational and fundamental stuff. However, applying lean manufacturing techniques to the “white collar factory” of sourcing and P2P (Purchase-to-Pay) is a mostly untapped area of opportunity.
Interestingly, some procurement organizations have named themselves “Supply Chain Management” even in non-manufacturing environments (e.g., Bank of America), but yet they always haven’t taken to heart key practices that manufacturing organizations have put in place on the shop floor. This is unfortunate, because it can be done.

Managing the “sourcing factory”

One way to view strategic sourcing is that of a Configure-to-Order business that “manufactures” highly profitable services. How profitable? For every $1 invested in procurement, world-class procurement delivers $7 to the firm, and that number goes even higher when looking at strategic supply processes. Unfortunately, there is a backlog of work because there is not enough investment in the bottleneck work centers (e.g., commodity managers), and not enough profitable services are getting out the door. So, attacking the bottleneck is critical, but funds are not unlimited to purchase more capacity, and must be freed from other areas (e.g., transactional processes) while improving “yield” through better work methods and measured doses of appropriate automation (e.g., freeing up commodity manager’s time via better spend/supplier analytics).

Another issue within the sourcing factory is aligning capacity to customer expectations via a “Capable-to-Promise” model. Various types of standard sourcing services, and their associated lead times and quality levels, should be offered up to customers based on finite capacity, and then configured to order. Without segmented “flow lines” (e.g., simple negotiations versus complex ones), standard lead times, capacity planning, and demand management (e.g., setting rules by which procurement must be involved in sourcing), the factory is going to be backlogged, quality will suffer, and customers will be very unhappy.

Designing what you can manufacture

The end of the sourcing factory is not the contract. A sourcing service is only profitable when preferred agreements are actually utilized within the “P2P factory” (where orders are placed and bills are paid). Unfortunately, they often aren’t. For the average company, our benchmark data puts overall bypass/maverick spend at 10%; but the real problem lies within indirect spend. A custom study that we did with 200 firms on contract management revealed a 23% maverick spending figure for influenced indirect spend. This translates to $11 million in lost savings per billion in indirect spend for the typical company. The problem with this $11 million of “scrap” is that the design of the Source-to-Settle process didn’t adequately consider the downstream processes of P2P (or supplier management and development). Strangely, every strategic sourcing methodology includes a “stakeholder management” process, yet the methodologies rarely explicitly define how P2P processes and systems will guide users to preferred supply sources and optimal buy/pay methods. It’s important to make strategic sourcing staff accountable for maverick spending (and not just savings). Treat P2P process users as customers – key stakeholders – and utilize thoughtfully designed downstream processes such as P2P and supplier management and development.

Converting the P2P job shop to flow lines

Most companies claim they have a defined P2P process, but if you scratch the veneer, you’ll find issues — e.g., only one-quarter of typical firms have single accountable P2P process owners. Frankly, some companies’ P2P processes are positively medieval, with each transaction handcrafted in a manner befitting the purchaser trying to get it through the system. If a firm has moved into the industrial age of P2P manufacturing and does have any P2P methods defined, it is likely the venerable three-way match. In manufacturing vernacular, this is known as a “job shop” — a “one facility fits all” general purpose processing capability, where everything goes in on one side and hopefully makes it out the other. If it’s an ERP environment, it’s “one system fits all”. In Lean manufacturing environments, flow lines (or “cells”) are set up based on families of similarly-made parts; for P2P processing, firms should define tailored transactional flow lines beyond the 3-way match, to include p-cards, assumed receipts, Evaluated Receipts Settlement (ERS), invoice-to-contract matching when POs not required, etc.

Papers from Hackett’s Purchase-to-Pay advisory program describe these concepts: “”Using an Optimized Transaction Strategy to Achieve P2P Efficiency”” and “A Management Primer for Balancing Risk and Control in P2P”. By designing a “P2P manufacturing” factory with transactional flow-lines that are fit-for-purpose, efficiency and effectiveness will invariably improve.

Thanks, Pierre!